How AI can make sense of a virus

The coronavirus. Surely, everyone has heard enough about it by now. But the reality is that it is unfortunately still a relevant topic. Therefore we, as a society, need to deal with it and try to minimize the impact that it has on all our lives. Apart from staying home, keeping your distance from others, and using your common sense, what else can be done? Since some of you might have something to do with Artificial Intelligence (AI), it might be worth asking if there are ways that this technology can help. And if so, how? The answer -what a surprise- is yes. There are many initiatives that use their ingenuity to process, interpret and act upon the large streams of data that are generated about the virus. Two of those are the CLAIRE task force on COVID19 from the Confederation of Laboratories for Artificial Intelligence Research in Europe (CLAIRE) and the Virus Outbreak Data Network (VODAN). What are these institutions and what do they do?

VODAN

Let us start with VODAN. VODAN was started in Leiden. It does not work with AI itself, but it plays an important role for those who do. It tries to make the available data as accessible and clear as possible. In their own words, they want to make the data FAIR (Findable, Accessible, Interoperable, and thus Reusable). This is to make sure that every country dealing with the crisis has access to possible life saving data. Not only for this particular crisis but also for future infectious disease outbreaks.

The goals of VODAN are to make sure that the gathered data is easy to read and that it is FAIR, create a system that makes it possible to easily insert new data (this data mostly comes from hospitals and general practitioners), assist in affected regions by transforming their data to FAIR data using local experts, install local data collection points, demonstrate the usefulness of the method and hopefully get WHO approval and lastly, offer the collected data to researchers who want to use it.

CLAIRE

CLAIRE is an organization that is backed by 35 countries with a total of 21,000 employees. It was started to present a united European front in the field of Artificial Intelligence. By combining these forces, a lot more work can be done. That is also why CLAIRE set up a special branch for combatting the virus. By bringing together the people who can handle data and the people who can handle what needs to be done based on the conclusions of the data, the process gets streamlined.

These decisions might have to do with the infectability of children, if it turns out that the data supports the claim that children hardly spread the virus, governments might consider removing restrictions for children. It could also work the other way around, if it turns out that there are many infections in a certain sector, it may be decided to temporarily set heavier restrictions.  Lastly, it could also predict the pressure on hospitals. If we can see that there are many patients expected, the upscaling process could be started earlier.

After the pandemic has passed, AI could still help. Since this virus will leave a lasting mark, AI could streamline the reconstruction of the economy by looking at what measures will be the most effective to get the economy running again. This can for example be used for companies that have logistic problems.

What can you do?

If you have a background in AI or in a related field and if you are now left wondering if you can contribute to one of these initiatives, do not hesitate to visit either the claire-ai.org or go-fair.org for more information about these projects.

For now: stay home, stay safe.

Authors:

Ruben Koole and the Symposium Committee of Study Association Flow, Tilburg, The Netherlands.

Sources:

Home

https://covid19.claire-ai.org

https://www.universiteitleiden.nl/nieuws/2020/03/bestrijding-corona-begint-met-het-verantwoord-delen-van-data

https://www.universiteitleiden.nl/nieuws/2020/04/ai-tegen-corona https://www.go-fair.org/implementation-networks/overview/vodan

Leave a Comment

Your email address will not be published. Required fields are marked *